Add live candidate queue checks#166
Conversation
Co-Authored-By: Codex <noreply@openai.com>
|
Bugbot is not enabled for your account, so this pull request was not reviewed. Enable Bugbot in the Cursor dashboard to get automatic reviews on future PRs. |
🤖 Codex PR ReviewPlease ensure a human reviewer checks this PR before merging. |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 6bdf246401
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| return 1 | ||
|
|
||
| if args.live_candidate_queue: | ||
| print(json.dumps(build_live_candidate_queue(), ensure_ascii=False, indent=2, sort_keys=True)) |
There was a problem hiding this comment.
Build the queue from the loaded config
When --live-candidate-queue is used after editing platform-config.json but before regenerating strategy-profiles.example.json, this calls build_live_candidate_queue() with no catalog, so it falls back to the checked-in generated catalog instead of the just-loaded config. That makes the machine-readable queue omit newly added or promoted live_candidate / shadow_candidate profiles and can report stale gate values even though the command validated the new config; pass the loaded strategies (or entries generated from them) so the queue reflects the canonical source.
Useful? React with 👍 / 👎.
Adds stricter default profile gate drift checks and a machine-readable live-candidate queue command.\n\nValidation: python3 -m unittest discover -s python/tests -v (83 passed), build_config --check passed, worker validation passed.